DI
SC
US
SI
ON
P
AP
ER
S
ER
IE
S
Forschungsinstitut zur Zukunft der ArbeitInstitute for the Study of Labor
Does Migration Make You Happy?A Longitudinal Study of Internal Migrationand Subjective Well-Being
IZA DP No. 6140
November 2011
Beata NowokMaarten van HamAllan M. FindlayVernon Gayle
Does Migration Make You Happy? A Longitudinal Study of Internal
Migration and Subjective Well-Being
Beata Nowok University of St Andrews
Maarten van Ham
Delft University of Technology and IZA
Allan M. Findlay University of St Andrews
Vernon Gayle University of Stirling
Discussion Paper No. 6140 November 2011
IZA
P.O. Box 7240 53072 Bonn
Germany
Phone: +49-228-3894-0 Fax: +49-228-3894-180
E-mail: [email protected]
Any opinions expressed here are those of the author(s) and not those of IZA. Research published in this series may include views on policy, but the institute itself takes no institutional policy positions. The Institute for the Study of Labor (IZA) in Bonn is a local and virtual international research center and a place of communication between science, politics and business. IZA is an independent nonprofit organization supported by Deutsche Post Foundation. The center is associated with the University of Bonn and offers a stimulating research environment through its international network, workshops and conferences, data service, project support, research visits and doctoral program. IZA engages in (i) original and internationally competitive research in all fields of labor economics, (ii) development of policy concepts, and (iii) dissemination of research results and concepts to the interested public. IZA Discussion Papers often represent preliminary work and are circulated to encourage discussion. Citation of such a paper should account for its provisional character. A revised version may be available directly from the author.
IZA Discussion Paper No. 6140 November 2011
ABSTRACT
Does Migration Make You Happy? A Longitudinal Study of Internal Migration and Subjective Well-Being
The majority of modelling studies on consequences of internal migration focus almost exclusively on the labour market outcomes and the material well-being of migrants. We investigate whether individuals who migrate within the UK become happier after the move than they were before it and whether the effect is permanent or transient. Using life satisfaction responses from 12 waves of the British Household Panel Survey (BHPS) and employing a fixed-effects model, we derive a temporal pattern of migrants’ subjective well-being (SWB) around the time of the migration event. Our findings make an original contribution by revealing for the first time that, on average, migration is preceded by a period when individuals experience a significant decline in happiness. The boost that is received through migration appears to bring people back to their initial level of happiness. As opposed to labour market outcomes of migration, SWB outcomes do not differ significantly between men and women. Perhaps surprisingly, long-distance migrants are at least as happy as short-distance migrants despite the higher social costs that are involved. JEL Classification: J61, R23 Keywords: migration, happiness, subjective well-being, longitudinal data, UK Corresponding author: Beata Nowok Centre for Population Change School of Geography and Geosciences University of St Andrews KY16 9AL St Andrews, Fife Scotland, UK E-mail: [email protected]
1
1. Introduction
The study of internal migration has been predominately shaped by economic questions
related to the labour market. At a micro-level, a decision to migrate is mainly explained by
cost-benefit calculations leading to an expected positive net return, usually measured by
monetary income. In other words, people are often assumed to maximize their utility
expressed in pecuniary terms (Chiswick, 2008). The migration decision-making process at
the family level is based on the evaluation of a joint return with potentially unequal gains or
losses for spouses (Cooke, 2008). Much empirical evidence indicates that migration is
beneficial for the careers of migrants, unless they are tied-movers (mostly women) who
sacrifice their labour-market outcome for the sake of the net gain to the family as a whole.
Compared with the economic consequences of migration, much less is known about
implications of migration for the general subjective well-being (SWB) of individuals.
Whether migration brings happiness to all migrants is a crucial question, especially in the
context of the growing interest in using well-being measures to evaluate societal progress. As
is common in the literature, we use the terms SWB and happiness interchangeably. They refer
to people’s evaluations of their life which can be cognitive or affective (e.g. Diener, 2009;
Diener et al, 2003). From a theoretical perspective, migration decision-making can still be
encompassed within a utility-maximizing framework. Now, however, utility is captured by
subjective judgments of satisfaction rather than monetary income.
A SWB framework for the analysis of migration has a number of advantages. It takes
a wider perspective that incorporates the richness and diversity of current geographical
mobility in terms of motivations and outcomes. People migrate for a variety of reasons, not
only economic ones, but most expect to increase their quality of life and happiness through
changing their place of residence. Migration is a stressful event requiring many adjustments
(Magdol, 2002; McCollum, 1990) and therefore it has to offer the chance to gain something
in return. Due to its complexity, migration affects many domains of the migrant life
regardless of the motivations of move, and these effects can be either positive or negative. An
outcome of migration on labour market does not have to coincide with the actual experience
of the quality of life in a destination area, especially in the context of the illusory general
belief that money brings happiness (Kahneman et al, 2006). Successful migration should,
however, reach its goal and increase the migrants’ subjective utility. We may expect negative
effects of migration on SWB when a person moves involuntarily or when he or she
mispredicts post-migration utility.
Evaluating migration results in terms of happiness rather than money has also
substantial policy implications (Diener and Seligman, 2004). Policies designed to stimulate
the economy by encouraging geographical mobility of households does not necessary lead to
increased individual and overall societal well-being. General well-being has been recognized
recently as a valuable measure to evaluate social progress and develop policy responses in
France (Stiglitz et al, 2009) and in the UK (Prime Minister’s Office, 25 November 2010;
Stratton, 2010).
This study advances the idea of exploring the relationship between migration and
SWB. We investigate whether individuals who migrate within the UK become happier after
the move than they were before it. For those who report higher levels of happiness after
migration, it is important to determine whether it is a permanent or transient effect. We use
panel data and observe levels of SWB both prior and after migration events. To effectively
follow happiness level of the same individuals over time we apply a fixed-effects panel data
2
models. Additionally, we are interested in whether or not the effects are different for various
types of migration.
The paper is structured as follows. Section 2 presents the background and previous
literature. Section 3 describes the data and method used in the study. Section 4 presents the
results of the empirical analysis. In Section 5 we present our conclusions.
2. Background
Set-point theory of SWB has been a prevailing paradigm in psychology (see Headey, 2010
for an overview; Lucas et al, 2003). The central premise is that individuals have stable levels
of SWB given by genetics and personality. Deviations from the set-points may occur in the
face of major life events, such as marriage, migration, unemployment, or serious injury, but
their effects are usually transitory. There is, however, increasing evidence of lasting changes
in individual happiness. Unemployment seems to be a reasonably common event which
causes long-term decrease in set-points (Clark et al, 2008a; Lucas et al, 2004). Lucas et al.
(2003) showed that marriage can have a long-lasting beneficial effect on SWB for some
people. Easterlin (2006a) concluded in his review article that in family and health domains
adaptation to changes is only partial, whereas people completely adapt to gains in the
economic domain. Headey (2010) found large and permanent changes in SWB set-points for
a large number of individuals participating in the German Socio-Economic Panel Survey
(SOEP). He challenged the set-point theory and called for a substantial revision. Considering
the emerging findings, it appears that happiness is shaped by both psychological factors and
life circumstances (Easterlin, 2006b). It implies that people can play an active role in
increasing their own happiness by making considered choices within their life strategies.
Migration can be seen as a means of potential lasting improvement in SWB. People
migrate for various reasons but most expect to improve their lot in one way or another. They
want to take advantage of opportunities available elsewhere. In terms of theoretical approach
human capital or cost-benefit models predominate in the migration literature. People are
assumed to behave rationally and move when the expected value of the benefits exceeds the
costs (pecuniary and non-pecuniary). Thus migration should be beneficial for migrants. When
more than one individual is involved in migration decision making, as in the case of family
migration, a rational evaluation of the cost and benefits of moving are often more complex.
Migration may be rational from the standpoint of the family as a whole, but not from the
standpoint of each family member considered separately. Family migration may therefore
provide disproportionate and unequal benefits to male and female partners (Coulter and van
Ham, 2012; Mulder and Cooke, 2009).
The economic gains that are conventionally taken in the literature as a suitable
measure of general migration outcome are one-sided and might be misleading. A key
question is whether migration is effective in raising happiness, not least in the long-term.
There are reasons to believe that this does not have to be the case. Migration is a move not
only in physical, but also in social space. It is a multifaceted event involving shifts in many
domains of life. An economically driven migration may be at considerable cost to social
relationships. Moreover, spatial mobility is closely and complexly interrelated with family
and career events which may not be neutral for SWB, e.g. divorce or loss of a job. An
individuals’ whole life satisfaction can be seen as an aggregate of satisfactions with the
various domains of their life (Cummins, 1996; Schimmack, 2008; van Praag et al, 2003).
However, the literature suggests that the weights assigned to various life domains may vary
by individuals and also they may change over the individual’s life span (McAdams et al,
2011; Pavot and Diener, in press). The studies adopting the life domain approach generally
3
agree on domains that are central to determining happiness. They include economic
conditions, family circumstances, health and work. All of these aspects of a person’s life may
be affected by a change of place of residence. Raising happiness in non-economic domains is,
however, more effective than in economic ones (Easterlin, 2006a). The transient effect of
income on life satisfaction is very often not realized by people (Kahneman et al, 2006).
As mentioned above, most research into the consequences of internal migration refers
to objective outcomes in the economic domain. Outcomes differ substantially between lead
and tied migrants. Women migrants are much more likely to be tied migrants than men
(Mincer, 1978), even when the woman actually is the primary wage earner (Cooke, 2003), or
has a higher ranking occupation than her partner (Boyle et al, 1999a). An extensive literature
demonstrates that following family migration women’s labour market status suffers in a
number of ways (Boyle et al, 1999b; 2001; Halfacree, 1995; Lichter, 1983; Mincer, 1978).
Women are less likely to be employed, have smaller incomes and work shorter hours than
other equivalent women (Boyle et al, 2001; Cooke and Bailey, 1999; Morrison and Lichter,
1988; Shihadeh, 1991). Studies using longitudinal data found, however, that married women
‘recover’ to pre-migration income levels within one to three years after the move (Clark and
Huang, 2006; LeClere and McLaughlin, 1997). The labour market outcome of migration is
usually positive for men when they are lead migrants.
Success in economic domain does not imply, however, an increase in happiness,
especially in the long term. The relation between income and happiness has attracted
widespread attention, especially among economists and research results reveal intricate
relationships (see e.g. Clark et al, 2008b; Easterlin, 1995; 2001; Easterlin et al, 2010; Ferrer-
i-Carbonell, 2005; Kahneman et al, 2006; Stutzer, 2004). There is a weak effect of absolute
income on SWB and several explanations emerge. First, in evaluating their financial situation
people compare themselves to others. As a result relative income rather than absolute income
is seen to affect happiness. Second, individuals adapt to material goods over time. People get
used most easily to material possessions and very often they underestimate the process of
habituation. Similarly, people’s expectations adapt to their possibilities. Finally, higher
income often shifts time-use towards activities associated with higher tension and stress.
Wealthier people tend to devote relatively more time to work and commuting and less time to
passive leisure activities (Kahneman et al, 2006).
The impact of other objective characteristics such as age, gender, marital status and
education on happiness level is surprisingly very limited. The contradictory evidence existing
in the literature prevents drawing firm conclusions for most of the examined correlates (for
reviews see Diener, 2009; Diener and Seligman, 2004; Dolan et al, 2008). We highlight the
influence of some selected migration-related characteristics. There is no significant difference
between men and women when average levels of SWB are considered. Females are more
likely, however, to report very high or very low levels of SWB. As regards age, many studies
have found that SWB in adulthood can be characterized by a U-shaped curve with the lowest
happiness occurring in middle age. A very useful overview of the relationship between well-
being and age is presented by Blanchflower and Oswald (2008).
Happiness research emphasizes that individual evaluations of well-being go beyond
economics. What is more important, raising happiness through material goods is almost
certainly doomed in the long run. We expect, therefore, that assessment of a long-term return
to migration in SWB terms will in most cases differ from calculations in monetary terms. An
important question, which we aim to answer, is whether migration has a positive impact on
individual long-term happiness level. Since happiness is considered by many as the ultimate
goal in life, voluntary migration should facilitate this goal. People may, however, exaggerate
the contribution of single factors to overall SWB, especially income. We expect also different
consequences of migration by gender for at least two reasons. First, men and women tend to
4
prioritize various domains of life that are affected differently by changing place of residence.
Second, women are more likely to be trailing spouses. Being tied-movers may be, however,
less harmful in SWB terms than in economic ones, since they may plausibly be less career-
oriented. Finally, there is no doubt that SWB return to migration depends on its broadly
defined type (short- versus long-distance migration).
3. Data and methods
3.1. Migration and the British Household Panel Survey
Panel data are especially beneficial in evaluation of the impact of migration as they allow a
comparison of the situation before and after the event. For this study we use the British
Household Panel Survey (BHPS), for the years 1996–2008 (this is the final year of the
original panel). In 1996 the questions on overall life satisfaction were introduced. The BHPS
is a nationally representative sample of approximately 5,500 private households with
approximately 10,000 adults recruited in 1991. The adult members (aged over 15) of the
same sample of households are interviewed every year. The following rules of the BHPS lead
to new households being included in the survey. The BHPS was also augmented by regional
geographical samples. Therefore in 2008 the total sample size was around 9,000 households
including some 15,000 individuals. The BHPS attempts to follow up all migrants who remain
in Great Britain. Although, as expected, attrition among migrants is higher than among non-
migrants, its extent is relatively small and does not pose a problem for the analysis of
geographical mobility (Buck, 2000; Rabe and Taylor, 2010).
The migration behaviour of the BHPS respondents can be tracked using two different
indicators provided in the dataset. Panel members are directly asked whether they still live at
the same residence as before 1 September of the previous year. In the case of change of
residence, information on month and year of the move is collected. The BHPS also provides a
derived individual mover status variable indicating whether sample members have moved
location since last interview. We identify migration events combining the information
contained in both variables. Migration is broadly defined as a change in the usual place of
residence (address) between two consecutive interviews. This includes both local and long-
distance moves. A migrant is a person who undertakes migration at least once during the
complete observation period (1996-2008). Depending on the number of migrations he or she
experiences we distinguish one-time migrants and multiple migrants. Our data indicate that,
on average, about 11% of people migrate every year. This adds up to a total of more than
20,500 migration events in the years 1996-2008. Around 38% of the migrating individuals
change place of residence more than once over the observation period. As a result there are
12,000 migrants in the dataset. It is common in migration research for analysts to consider
only the first move rather than all moves. The first migration observed in the dataset is not
necessarily the first migration experienced in the respondent’s life. Therefore, a focus on first
moves only would to some extent be artificial.
On average an individual moves a distance of 32 km. Nonetheless, half of moves are
of distance shorter than 3.2 km. About 9% of all moves are motivated by reasons related to
respondent’s job. The share of job-related migrations increases with distance - 31% for
distance exceeding 25 km as opposed to 3% below this threshold. The number of migrations
experienced by heterosexual couples (both married and cohabiting) equals about 3,600.
Approximately 11% of those couples migrate for reasons associated with job of either or both
partners. Half of the job-related moves of the couples are associated with only man’s job,
while around 20% with only woman’s job.
5
The SWB measure is derived by the evaluation question: ‘How dissatisfied or
satisfied are you with your life overall?’ There are seven possible response options ranging
from ‘not satisfied at all’ to ‘completely satisfied’. A neutral point of the scale is 4 at which
respondents report that they are neither satisfied nor dissatisfied. Despite its simplicity, this
single-item measure appears to be a fairly robust indicator of SWB and is frequently
employed in the literature. It is also used as a dependant variable in this study.
3.2. Modelling approach
Our objective is to examine how SWB changes relative to the time of migration. To account
for individual differences and effectively track the same people over time we apply a fixed-
effects model1. In order to capture the time path of SWB we create a series of dummy
duration variables. They represent the number of years before or after the occurrence of a
migration event. The resulting regression equation takes the form:
∑
, (1)
where denotes the subjective well-being of individual i in period t. The individual
fixed effect, , controls for any time-constant unobserved heterogeneity. is a vector of
time-varying covariates. It includes a set of individual and household characteristics that are
common in the literature on SWB. In particular, we consider age, marital status, labour
market status and self-assessed health at an individual level and birth of a child, number of
children and income at the household level. The dummy variables, , indicate if a person i
migrates in period t-k, with k indexing the variables beginning years before and ending
years after migration. The last group refers to all years beyond . For instance, if an
individual i migrated three years before year t. In other words, at year t he or she has been
living in a new place of residence for three years. If , it indicates that a person i will
migrate from a current place of residence in three years. The parameters measure therefore
the impact of migration prior to ( ) and following the move ( ). Finally, is a
stochastic error term, indexed i for the individual and t for time.
A similar modelling approach was adopted for examining life satisfaction effects of
major life events by Clark et al. (2008a) and Frijters et al. (2011). In economic literature
analogous models are used to evaluate earnings losses of displaced workers (Couch, 2001;
Couch et al, 2011; Couch and Placzek, 2010; Jacobson and LaLonde, 1993; White, 2010).
Given the nature of the dependant variable (seven ordered outcomes) an ordered
response regression might seem more appropriate. There are two main reasons in favour of a
linear model. First, linear analysis is superior in its ease of interpretation. Second, the analysis
of SWB carried out by the two models leads to similar substantive results (Clark et al, 2008a;
Ferrer-i-Carbonell and Frijters, 2004).
4. Results
4.1. Empirical regularities in SWB and migration
Our data confirm that most people are reasonably happy (Diener and Diener, 1996). Around
76% of the BHPS respondents indicate SWB above neutral. The SWB metric records an
average level of happiness of 5.23 and there is no significant difference between men and
1 We have also estimated the model in a random-effects framework and formally compared the models using the
Hausman test. The results favour a fixed-effects specification.
6
women. Migrants are, on average, significantly less happy than non-migrants. Migrants are
defined as in Section 3 and thus the aggregate measures for them are derived from all
available observations prior to and after migration(s). The average SWB scores for migrants
and non-migrants are respectively 5.16 and 5.30. Multiple migrants are again unhappier
(5.12) than one-time migrants (5.20).
SWB and migration exhibit strong regularities in age profiles (Figure 1). Happiness
tends to be approximately U-shaped over the adult life cycle with the minimum occurring in
middle age. The most prominent regularity in the age schedule of migration is the high
concentration of migration among young adults. The differences in aggregate measures of
happiness between migrants and non-migrants may, therefore, reflect a compositional effect
of age. The empirical age profiles of migration and SWB of both migrants and non-migrants
(together and separately) are set out in Figure 1.
Figure 1 Average subjective well-being (SWB) and migration rate by age, 1996-2008
We can make a few observations based on the presented age schedules and the relationship
between them. The average SWB of the total population starts high in young adulthood,
reaches the minimum at age 45, then rises substantially to age 70 and starts to decline
thereafter. Most migration events are concentrated between ages 18-30. Non-migrants are
happier that non-migrants at all ages except between 23 and 40 years. The higher SWB of
migrants at most mobile ages, which drives the happiness of the whole population, especially
at ages 23-30, suggest a positive impact of changing place of residence, at least in short term.
An increase in SWB for migrants may be also attributed to the relationship of age to some
aspects of the family life-cycle and the resulting motives for migration, e.g. marriage. In
order to understand the impact of migration on SWB, it is therefore of importance to follow
the same individuals over time, and control for other life events and personal characteristics.
Before we move on to the analysis of the long-term dynamics of SWB, we present
what we can learn from comparison of male and female happiness before and after migration
depending on distance and reason for the move (values can be consulted in Table 1). The
available data do not allow estimating long-term SWB patterns in small subgroups making
such descriptive analysis even more informative. At the first interview following migration
event the respondents’ average SWB is 5.15, which is higher by 0.05 than that at the
interview preceding the move. Distance itself has a limited impact on the happiness level
reported after the change of place of residence. Nonetheless, migrants who move a distance
7
of 25-50 km become happier after migration. As expected, reasons for moving influence the
migration outcome. In general, migrants motivated by job issues are happier than those
migrating for other reasons regardless of the migration distance.
Table 1 Post-migration subjective well-being (SWB) by reason, gender and distance.
Reason for
migration Gender
Distance in kilometres
0-5 5-10 10-25 25-50 50-100 100-200 200+
Total Total 5.15
(0.05) a
5.12
(0.05)
5.15
(0.11)
5.23
(0.16)
5.22
(-0.02)
5.13
(-0.02)
5.15
(0.13)
Job
Male 5.22
(0.16)
5.23
(0.37)
5.27
(0.39)
5.34
(0.25)
5.31
(0.21)
5.21
(0.16)
5.21
(0.15)
Female 5.36
(0.23)
5.33
(0.44)
5.15
(0.10)
5.29
(0.27)
5.25
(0.00)
5.18
(0.06)
5.17
(0.19)
Other
reasons
Male 5.14
(0.01)
5.15
(0.02)
5.12
(0.10)
5.16
(0.23)
5.15
(-0.05)
5.04
(-0.13)
5.12
(0.24)
Female 5.16
(0.08)
5.08
(0.04)
5.18
(0.11)
5.27
(0.06)
5.18
(-0.09)
5.03
(0.00)
5.03
(0.08)
Notes: a in parentheses change compared to pre-migration SWB.
The SWB of males and females does not differ considerably. There are, however, some
remarkable discrepancies between the happiness of partners, both married and cohabiting,
migrating together (Table 2). Men become happier only when migration is related to their
own job. It is even more rewarding when spouse’s job plays a role as a motivating factor as
well. The post-migration SWB of women migrating with partners remains practically the
same as it was before the move, regardless of migration reason. They are relatively happy and
being tied migrant is not really harmful for them.
Table 2 Post-migration subjective well-being (SWB) of partners in migrating couples a
Reason for
migration
No. of migrations
by couples Male Female
Man's job 218 5.35 ( 0.12)b 5.35 (-0.03)
Woman's job 96 5.07 (-0.06) 5.30 ( 0.00)
Both partners' jobs 133 5.39 ( 0.18) 5.27 ( 0.03)
Other reason 2841 5.21 (-0.01) 5.28 ( 0.01)
Notes: a couples include both married and cohabiting partners; b in parentheses change compared to pre-migration SWB.
4.2. Dynamic effect of migration on SWB
The examined period when migration is assumed to affect happiness spreads over nine years.
It begins four years prior to the year when a person reports a move and ends five years after
that year. The coefficient is, therefore, fixed and equal to zero (see Equation (1) in
Section 3). The estimated dynamic effect of migration on SWB for the full sample and for
men and women separately is illustrated in Figure 2. Here and in the following figures the
vertical line at zero indicates migration but, to be precise, time zero is the time of interview.
Inevitably the interview occurred after the migration event. Thus the migration act took place
at an unspecified time between minus one and zero on the graph. Detailed regression results
are presented in Table 3.
8
Figure 2 The dynamic effect of migration on subjective well-being (SWB) for total, males and females.
Notes: Error bars indicate 95% confidence intervals for total.
For the full sample, the coefficients on the migration-related dummies are all negative except
for the year of migration (Figure 2 and the first ten rows of Table 3). The SWB drops to a
level significantly lower than zero three years before the change of residence and remains
significantly negative till the year preceding the change. The first important finding of the
research is that migration boosts migrants’ happiness relative to people’s feelings of well-
being before moving. More precisely, migration seems to take away negative feelings or
unhappiness, but as Figure 2 shows the upward shift in subjective well-being does not
continue in the years after migration. In the year of migration the SWB seems to have
returned to its original level and stagnates thereafter. This pattern would be compatible with
an interpretation of migration taking place as a result of some stressor. Moving to
overcoming the stressor is therefore a positive action, but it should not be interpreted in
bringing any extra happiness or improved well-being relative to the migrant’s status before
the stressor took effect. Table 3 shows that the coefficients for the years after migration are
not significant. The interpretation of the findings as presented above would uphold research
on the ‘immobility paradox’ (Fischer and Malmberg, 2001) that notes that most people do not
move unless they have to, even when the measurable gains that mobility might offer are self-
evident. By contrast some people move because of unacceptable stress in one or more
domains of their lives reducing well-being to unacceptable levels. A final feature of Table 3
that seems worthy of note in relation to the coefficients relating to SWB in years before and
after migration is that the research would seem to substantiate the view that it is
circumstances at a time well before migration that may be critical in bringing about the
decision to move, and that the search process for a new location that satisfies the potential
movers concerns may take some time to find – hence the three year lag between the low point
on the SWB graph and the time of migration.
As regards the effects of the other variables on SWB, the results confirm a strong
relationship between SWB and self-assessed health and a large negative impact of
unemployment, which are usually found in the literature. Being married or living as a couple
appears to increase SWB. Being separated or widowed is associated with the lowest level of
happiness. Moreover, there is a positive effect of household income on the SWB of its
members.
9
Table 3 Fixed-effects model of subjective well-being; the coefficient estimates for the whole sample
Variable a Total Males Females
No. of years before and after
migration
-4 -0.037 (0.030) b -0.039 (0.042) -0.033 (0.041)
-3 -0.124 ***
(0.027) -0.108 ***
(0.039) -0.134 ***
(0.038)
-2 -0.079 ***
(0.024) -0.077 **
(0.034) -0.078 **
(0.033)
-1 -0.071 ***
(0.023) -0.063 * (0.033) -0.073
** (0.032)
0 0.011 (0.023) 0.004 (0.032) 0.023 (0.031)
1 -0.024 (0.024) -0.030 (0.034) -0.013 (0.033)
2 -0.016 (0.025) -0.011 (0.035) -0.016 (0.034)
3 -0.027 (0.026) 0.004 (0.036) -0.048 (0.036)
4 -0.019 (0.027) -0.014 (0.038) -0.020 (0.037)
5 -0.038 (0.026) -0.053 (0.037) -0.024 (0.036)
Age -0.007 ***
(0.003) -0.019 ***
(0.004) 0.002 (0.004)
Age squared/100 -0.001 (0.002) 0.011 ***
(0.004) -0.011 ***
(0.003)
Marital status
Married / living as couple 0.128 ***
(0.017) 0.118 ***
(0.025) 0.136 ***
(0.024)
Widowed -0.210 ***
(0.031) -0.144 ***
(0.051) -0.230 ***
(0.041)
Divorced -0.081 ***
(0.028) -0.141 ***
(0.044) -0.042 (0.038)
Separated -0.282 ***
(0.031) -0.395 ***
(0.048) -0.206 ***
(0.042)
Labour market status
Unemployed -0.305 ***
(0.017) -0.334 ***
(0.023) -0.278 ***
(0.026)
Student 0.110 ***
(0.018) 0.126 ***
(0.026) 0.102 ***
(0.024)
Long term sick, disabled -0.371 ***
(0.021) -0.448 ***
(0.030) -0.308 ***
(0.028)
Child born this year 0.077 ***
(0.015) 0.030 (0.022) 0.115 ***
(0.021)
Number of children -0.010 * (0.006) 0.007 (0.008) -0.027
*** (0.008)
Equivalized household income c
0.012 ***
(0.002) 0.017 ***
(0.003) 0.009 ***
(0.003)
Health status over last 12 months
Excellent 0.369 ***
(0.010) 0.363 ***
(0.015) 0.375 ***
(0.015)
Good 0.233 ***
(0.008) 0.240 ***
(0.011) 0.226 ***
(0.011)
Poor -0.281 ***
(0.012) -0.296 ***
(0.019) -0.273 ***
(0.016)
Very poor -0.645 ***
(0.022) -0.718 ***
(0.036) -0.604 ***
(0.029)
Notes: a coefficients on wave dummies are not reported; reference categories are never married and fair health; b standard
errors in parentheses; c in thousand pounds; * significant at 10%, **significant at 5%,*** significant at 1%.
There are no substantial gender differences in the impact of migration on SWB.
Similarly the research showed that females migrating in couples experience similar SWB
trends as their partners, although these results are not presented here. Given the literature on
the labour market consequences of migration, which indicate that migration is harmful for
females, this finding runs counter to the expectations. Women, who are tied migrants more
often than men, do not seem to sacrifice their happiness while losing their career
opportunities. It may suggest that women attach lower importance to work than men. It
should also be noted that men’s happiness appears not to be affected either by the birth of a
child or the number of children in the household.
The great diversity of population mobility means that it would be desirable to
undertake an analysis disaggregating movers by their self-defined motivations for migration
as well as by types of move in terms of distance and household type. An elaborate division of
10
migration events into subgroups would lead, however, to seriously limited sample sizes. Thus
we restrict the analysis here to just a broad categorisation of movement types. We distinguish
local moves from distant moves with a threshold set at 25 km (dominated by moves for
housing and personal reasons and excluding longer distance moves that tend to be rather
different in nature often being associated with labour market switchers as well as migration
for educational purposes, especially the migration of students from their parental home to
study at universities). This threshold also approximates to the significant divide recognised in
migration studies between moves that permit the mover to maintain social ties with most of
their former social network and longer distance moves that require the migrant to form new
social bonds. On this basis one might anticipate that short distance movers would be happier
than longer distance movers simply because they experience less social costs in moving,
since their social network is not disrupted to the same extent.
Figure 3 plots changes in the SWB around the time of the migration event estimated
separately for long- and short-distance moves. Curiously, and counter to the hypothesis
established above about mobility, well-being and the breaking of social networks, long-
distance migrants seem to be happier in general than the short-distance movers. Moreover,
the positive effect of migration is observed for a longer period (two years after move) before
it starts to wear off. The sample size for longer distance migrants is small, however, and so
some caution is required to limit reading too much into trends for this group. We focus,
therefore, on residential mobility only. The coefficients and the standard errors for the model
estimated for men and women together and separately are presented in Table 4. Figure 3 and
Figure 4 illustrates the general shape of changes in SWB as a function of short-distance
migration for the same groups.
Figure 3 The dynamic effect of short- and long-distance migration on subjective well-being (SWB)
Notes: Error bars indicate 95% confidence intervals for short- and long-distance migrations.
Interestingly, local moves take place after a relatively bigger decline in SWB and movers do
not recover to their initial level of well being after their move although, as with Figure 2, the
SWB graph stagnates in the years after the move has taken place. The effect of migration
distance is driven by changes in happiness experienced by men. In the case of women, its
impact is negligible. Again, it suggests that relative importance of life domains in
determining general life satisfaction differs between genders. Males get lower satisfaction
from changes related to residential mobility. Migration to a distant place that usually offers
them more career opportunities and a chance to begin a new life appears to be more effective
in boosting their happiness, but as Figure 3 shows, this effect is not quite statistically
significant.
11
Figure 4 The dynamic effect of short-distance migration on subjective well-being (SWB) of males (left panel)
and females (right panel).
Notes: Error bars indicate 95% confidence intervals for short-distance migrations.
The differences in the effects of personal characteristics on the SWB of people migrating
over various distances are limited: short-distance movers are more negatively affected by
very poor health. Their happiness is not significantly influenced by being in the widowed
category.
Table 4 Fixed-effects model of subjective well-being; the coefficient estimates for the short-distance migrations
Variable a Total Males Females
No. of years before and after
migration
-4 -0.079 **
(0.03) b -0.089
* (0.049) -0.066 (0.049)
-3 -0.136 ***
(0.027) -0.133 ***
(0.046) -0.131 ***
(0.045)
-2 -0.119 ***
(0.024) -0.146 ***
(0.041) -0.092 **
(0.04)
-1 -0.094 ***
(0.028) -0.112 ***
(0.04) -0.071 * (0.04)
0 -0.023 (0.028) -0.069 * (0.04) 0.024 (0.04)
1 -0.069 **
(0.03) -0.116 ***
(0.044) -0.023 (0.042)
2 -0.053 * (0.032) -0.078
* (0.045) -0.026 (0.044)
3 -0.048 (0.033) -0.049 (0.047) -0.040 (0.046)
4 -0.050 (0.035) -0.105 **
(0.049) -0.001 (0.048)
5 -0.060 (0.037) -0.130 **
(0.052) 0.000 (0.051)
Age -0.007 * (0.004) -0.025
*** (0.006) 0.007 (0.006)
Age squared/100 0.000 (0.004) 0.022 ***
(0.007) -0.016 ***
(0.006)
Marital status
Married / living as couple 0.113 ***
(0.022) 0.127 ***
(0.032) 0.107 ***
(0.03)
Widowed -0.096 * (0.055) -0.065 (0.092) -0.105 (0.069)
Divorced -0.072 * (0.037) -0.110
* (0.057) -0.045 (0.05)
Separated -0.247 ***
(0.04) -0.311 ***
(0.061) -0.195 ***
(0.054)
Labour market status
Unemployed -0.337 ***
(0.027) -0.376 ***
(0.037) -0.307 ***
(0.04)
Student 0.115 ***
(0.029) 0.118 ***
(0.045) 0.118 ***
(0.038)
Long term sick, disabled -0.404 ***
(0.034) -0.493 ***
(0.049) -0.333 ***
(0.047)
Child born this year 0.104 ***
(0.02) 0.065 **
(0.029) 0.138 ***
(0.028)
Number of children -0.013 (0.008) 0.005 (0.011) -0.032 ***
(0.011)
Equivalized household income c 0.018
*** (0.004) 0.025
*** (0.005) 0.013
** (0.005)
12
Health status over last 12 months
Excellent 0.395 ***
(0.017) 0.365 ***
(0.024) 0.420 ***
(0.023)
Good 0.243 ***
(0.013) 0.239 ***
(0.019) 0.244 ***
(0.018)
Poor -0.302 ***
(0.021) -0.329 ***
(0.033) -0.289 ***
(0.027)
Very poor -0.742 ***
(0.038) -0.870 ***
(0.06) -0.671 ***
(0.048)
Notes: a coefficients on wave dummies are not reported; reference categories are never married and fair health; b standard
errors in parentheses; c in thousand pounds; * significant at 10%, **significant at 5%,*** significant at 1%.
5. Discussion and conclusion
The overall goal of this paper has been to answer the question ‘Does migration make you
happy?’ It does so using contemporary longitudinal data from the BHPS. The use of panel
data allowed the analysis to be innovative in examining whether the effects of migration on
happiness are transient or permanent. The longitudinal analysis of changes in SWB around
the time of migration event presented in this paper provides new insights into implications of
migration. Previous studies have mostly focused on the labour market consequences of
migration and have neglected other aspects of social life. There are significant SWB changes
associated with mobility with the strongest effect in the year of migration. Migrants are
happier just after the move than they were just prior to it. A broader temporal perspective
reveals, however, that migration is preceded by a decline in SWB. Opportunities available to
migrants at new places of residence seem to provide a way out of unhappiness. An alternative
hypothesis could be that a decline in happiness before the move reflects the anticipation of
the negative effects of moving (in relation to the associated stresses of social networks
disruption, leaving familiar surroundings and adjusting to the new environment). The
research suggests that a boost in happiness is received through migration bringing people
back to their initial level of SWB (Figure 2). Counter to our expectations, long-distance
movers are at least as happy as short-distance movers despite the higher social costs that are
involved. Moreover, happiness outcomes after migration, as opposed to labour market
outcome, do not differ significantly by gender. The happiness of women, who are more often
tied migrants, does not seem to be dented even though their career opportunities may become
more limited.
In the broader context of SWB our results support the set-point theory of happiness.
Individuals shift away from a baseline SWB but they tend to come back to their long-term
happiness level. The questionable possibility of lasting improvements in SWB also feeds into
the ongoing debate on policy aiming at improving the well-being of societies.
Finally, note that since SWB includes an important temporal dimension, analysing
long-run longitudinal data is essential for research to offer meaningful insights into the
drivers of and barriers to happiness. Analyses exploring happiness at only one point of time
or comparing levels of SWB only before and after the migration event (i.e. at levels reported
in only two panel waves before and after migration) , are likely to arrive at erroneous
conclusions.
Despite its contribution to the understanding of both migration and happiness our
study has a few limitations. Although BHPS data enable us to link migrating individuals
within couples and to identify lead and tied spouses, the resulting sample size is too small to
draw reliable conclusions on long-term happiness patterns. Our findings on tied movers are
based, therefore, on comparison between men and women and an underlying assumption that
most trailing spouses are women. In addition, the relative importance of various life domains
for males and females in migration context needs further investigation. Data-related problems
limit also the analysis by migration type. It would be desirable to investigate the happiness
13
consequences of migration by its motives. We distinguish short-distance moves from long-
distance moves but it is only a crude approximation of migration reasons. Our research rests
on the plausible assumption that measures of migration will be exogenous to SWB. This
assumption could in theory be explored further through a simultaneous equations approach.
However this approach is not trivial and in our experience the relevant set of explanatory
variables is not available within the BHPS.
In closing, we would ask and attempt to briefly answer two questions. First, what has
this paper contributed to the wider understanding of migration and well-being? Second, what
are the implications of this research in relation to exploring the relations between migration
and set-point theory?
In answer to the first question, the paper has sought to offer a particularly rigorous
analysis of one metric associated with well-being relative to the timing of migration. The
originality of the findings, as summarised above, lies in showing that for the individual mover
there is relationship with happiness that is a transient one, relating primarily to the years
before migration. The returns to migration in terms of happiness appear to be time- specific
and not to accumulate after migration, unlike for example potential returns to human capital
where one would expect that benefits might accrue to the migrant over many years (in terms
for example of increased earnings) as anticipated by neo-classical models of labour
migration. Those seeking to theorise migration at the scale of the individual mover need
therefore to consider why non-economic returns as measured by indicators of well-being
suggest only transient losses and gains. The research findings might therefore be taken to
challenge extant theorisations of migration, such as those suggesting that the deeper drivers
of migration lie in the social and cultural meanings of mobility rather than in more easily
understood economic rewards (Findlay and Stockdale, 2003; Halfacree and Boyle, 1993). A
more likely resolution of the differences between the authors’ findings and the
conceptualisation of migration in the research literature might be found in the suggestion that
the research results in this paper relate specifically to short-distance internal migration
(involving moves associated with factors like housing needs and life-course adjustments such
as divorce (Boyle et al, 2008)), and that different findings might emerge from analysis of
well-being before and after international migration. In the latter case different scales of
cultural and social disruption seem probable that would be anticipated to be linked to longer
run migrant experiences of integration or exclusion. This then emerges as a key research
agenda that could usefully be tackled using other datasets.
Turning finally to the wider implications in relation to set-point theory, some might
suggest that the findings indicate that internal migration is unlike life events such as
marriage, loss of a job or major injury. While they often have a lasting impact on happiness
(Clark et al, 2008a; Lucas et al, 2003), internal migration could be interpreted from the results
reported in this paper to confirm set point theory views that individuals have stable levels of
well-being and that once destabilising events such as migration (or the causes necessitating
migration) have been negotiated, then people return to their original level of well-being. An
alternative interpretation (and one that we believe our results begin to support as a result of
offering detailed insights of temporal variations in SWB) would be to argue that the prospects
of mobility as well as the act of engaging in internal migration may be critical means for
restoring an individual’s level of social well-being, especially following previous stressful
events. Indeed one might suggest that without migration the stable sense of well-being
anticipated by set point theory might not be regained. Resolving this quandary may well be
possible through further research linking motivations for migration to models of well-being in
relation to specific life domains. For example, it would be logical to anticipate that features of
life satisfaction, such as employment, would be affected very differently as a consequence of
migration from domains, such as use of leisure time. More detailed analysis of the relations
14
between social well-being and migration, might therefore help not so much to answer the
question ’does migration make you happy?’, but rather to substantiate the claim that mobility
is one of several means by which an individual can regain a stable sense of well-being in the
fashion anticipated by set point theory.
References
Blanchflower D G, Oswald A J, 2008, "Is well-being U-shaped over the life cycle?" Social Science and
Medicine 66 1733-1749
Boyle P, Cooke T, Halfacree K H, Smith D, 1999a, "Gender inequality in employment status following family
migration in GB and the US: the effect of relative occupational status" International Journal of Sociology
and Social Policy 19 109-143
Boyle P, Cooke T, Halfacree K H, Smith D, 1999b, "Integrating GB and US Census Microdata for studying the
impact of family migration on partnered women's labour market status" International Journal of Population
Geography 5 157-178
Boyle P, Cooke T, Halfacree K H, Smith D, 2001, "A cross-national comparison of the impact of family
migration on women's employment status" Demography 38 201-213
Boyle P, Kulu H, Cooke T, Gayle V, Mulder C H, 2008, "Moving and union dissolution" Demography 45 209-
22
Buck N H, 2000, "Using panel surveys to study migration and residential mobility", in Researching Social and
Economic Change: The Uses of Household Panel Studies Ed. D Rose (Routledge, London) pp 250-272
Chiswick B R, 2008, "Are immigrants favorably self-selected? An economic analysis", in Migration Theory:
Talking Across Disciplines Eds C D Brettell, J F Hollifield 2nd edition (Routledge, New York) pp 63-82
Clark A E, Diener E, Georgellis Y, Lucas R E, 2008a, "Lags and leads in life satisfaction: a test of the baseline
hypothesis" Economic Journal 118 F222-F243
Clark A E, Frijters P, Shields M A, 2008b, "Relative income, happiness, and utility: an explanation for the
Easterlin paradox and other puzzles" Journal of Economic Literature 46 95-144
Clark W A V, Huang Y, 2006, "Balancing move and work: women's labour market exits and entries after family
migration" Population, Space and Place 12 31-44
Cooke T J, 2003, "Family migration and the relative earnings of husbands and wives" Annals of the Association
of American Geographers 93 338-349
Cooke T J, 2008, "Gender role beliefs and family migration" Population, Space and Place 14 163-175
Cooke T J, Bailey A J, 1999, "The effect of family migration, migration history and self-selection on married
women's labour market achievement", in Migration and Gender in the Developed World Eds P Boyle, K
Halfacree (Routledge, London) pp 102–113
Couch K A, 2001, "Earnings losses and unemployment of displaced workers in Germany" Industrial and Labor
Relations Review 54 559-572
Couch K A, Jolly N A, Placzek D W, 2011, "Earnings losses of displaced workers and the business cycle: an
analysis with administrative data" Economics Letters 111 16-19
Couch K A, Placzek D W, 2010, "Earnings losses of displaced workers revisited" American Economic Review
100 572-589
Coulter R, van Ham M, 2012, "Partner disagreement on moving decisions and the subsequent moving behaviour
of couples" Population Space and Place (forthcoming)
Cummins R A, 1996, "The domains of life satisfaction: An attempt to order chaos" Social Indicators Research
38 303-328
Diener E, 2009, "Subjective well-being", in The Science of Well-Being: The Collected Works of Ed Diener Ed. E
Diener (Springer, The Netherlands) pp 11-58
Diener E, Diener C, 1996, "Most people are happy" Psychological Science 7 181-185
Diener E, Scollon C N, Lucas R E, 2003, "The evolving concept of subjective well-being: the multifaceted
nature of happiness" Advances in Cell Aging and Gerontology 15 187-219
Diener E, Seligman M E P, 2004, "Beyond money: toward an economy of well-being" Psychological Science in
the Public Interest 5 1-31
Dolan P, Peasgood T, White M, 2008, "Do we really know what makes us happy? A review of the economic
literature on the factors associated with subjective well-being" Journal of Economic Psychology 29 94-122
Easterlin R A, 1995, "Will raising the incomes of all increase the happiness of all?" Journal of Economic
Behavior and Organization 27 35-47
Easterlin R A, 2001, "Income and happiness: towards a unified theory" The Economic Journal 111 465-484
15
Easterlin R A, 2006a, "Building a better theory of well-being", in Economics and Happiness: Framing the
Analysis Eds L Bruni, P Porta (Oxford University Press, Oxford)
Easterlin R A, 2006b, "Life cycle happiness and its sources. Intersections of psychology, economics, and
demography" Journal of Economic Psychology 27 463-482
Easterlin R A, McVey L A, Switek M, Sawangfa O, Zweig J S, 2010, "The happiness–income paradox
revisited" Proceedings of the National Academy of Sciences 107 22463-22468
Ferrer-i-Carbonell A, 2005, "Income and well-being: An empirical analysis of the comparison income effect"
Journal of Public Economics 89 997-1019
Ferrer-i-Carbonell A, Frijters P, 2004, "How important is methodology for the estimates of the determinants of
happiness?" Economic Journal 114 641-659
Findlay A M, Stockdale A, 2003, "The temporal and social embeddedness of migration: a methodological
exploration using biographical analysis" Geography Research Forum 23 4-29
Fischer P A, Malmberg G, 2001, "Settled people don't move: on life course and (im-)mobility in Sweden"
International Journal of Population Geography 7 357-371
Frijters P, Johnston D W, Shields M A, 2011, "Life satisfaction dynamics with quarterly life event data"
Scandinavian Journal of Economics 113 190-211
Halfacree K H, 1995, "Household migration and the structuration of patriarchy: evidence from the USA"
Progress in Human Geography 19 159-182
Halfacree K H, Boyle P, 1993, "The challenge facing migration research: the case for a biographical approach"
Progress in Human Geography 17 333-348
Headey B, 2010, "The set point theory of well-being has serious flaws: on the eve of a scientific revolution?"
Social Indicators Research 97 7-21
Jacobson L S, LaLonde R J, 1993, "Earnings losses of displaced workers" American Economic Review 83 685
Kahneman D, Krueger A B, Schkade D, Schwarz N, Stone A A, 2006, "Would you be happier if you were
richer? A focusing illusion" Science 312 1908-1910
LeClere F B, McLaughlin D K, 1997, "Family migration and changes in women's earnings: a decomposition
analysis" Population Research and Policy Review 16 315-335
Lichter D T, 1983, "Socioeconomic returns to migration among married women" Social Forces 62 487-503
Lucas R E, Clark A E, Georgellis Y, Diener E, 2003, "Reexamining adaptation and the set point model of
happiness: reactions to changes in marital status" Journal of Personality and Social Psychology 84 527-539
Lucas R E, Clark A E, Georgellis Y, Diener E, 2004, "Unemployment alters the set point for life satisfaction"
Psychological Science 15 8-13
Magdol L, 2002, "Is moving gendered? The effects of residential mobility on the psychological well-being of
men and women" Sex Roles 47 553-560
McAdams K K, Lucas R E, Donnellan M B, 2011, "The role of domain satisfaction in explaining the
paradoxical association between life satisfaction and age" Social Indicators Research (forthcoming)
McCollum A, 1990 The Trauma of Moving: Psychological Issues for Women (Sage Publications, London)
Mincer J, 1978, "Family migration decisions" The Journal of Political Economy 86 749-773
Morrison D R, Lichter D T, 1988, "Family migration and female employment: the problem of underemployment
among migrant married women" Journal of Marriage and Family 50 161-172
Mulder C H, Cooke T J, 2009, "Family ties and residential locations" Population, Space and Place 15 299-304
Pavot W, Diener E, in press, "Happiness experienced: The science of subjective well-being", in Oxford
Handbook of Happiness Eds I Boniwell, S David (Oxford University Press, Oxford)
Prime Minister’s Office, 25 November 2010, "PM speech on wellbeing",
http://www.number10.gov.uk/news/speeches-and-transcripts/2010/11/pm-speech-on-well-being-57569
Rabe B, Taylor M, 2010, "Residential mobility, quality of neighbourhood and life course events" Journal of the
Royal Statistical Society: Series A (Statistics in Society) 173 531-555
Schimmack U, 2008, "The structure of subjective well-being", in The Science of Subjective Well-Being Eds R
Larsen, M Eid (Guilford Press, New York) pp 97–123
Shihadeh E S, 1991, "The prevalence of husband-centered migration: employment consequences for married
mothers" Journal of Marriage and Family 53 432-444
Stiglitz J, Sen A, Fitoussi J, 2009, "Report by the Commission on the Measurement of Economic Performance
and Social Progress", www.stiglitz-sen-fitoussi.fr
Stratton A, 2010, "David Cameron aims to make happiness the new GDP" The Guardian 15 November, page
14, http://www.guardian.co.uk/politics/2010/nov/14/david-cameron-wellbeing-inquiry
Stutzer A, 2004, "The role of income aspirations in individual happiness" Journal of Economic Behavior &
Organization 54 89-109
van Praag B M S, Frijters P, Ferrer-i-Carbonell A, 2003, "The anatomy of subjective well-being" Journal of
Economic Behavior & Organization 51 29-49
White R, 2010, "Long-run wage and earnings losses of displaced workers" Applied Economics 42 1845-1856